Low Dimensional Motor Skill Learning Using Coactivation

Ranganath, Avinash; Xu, Pei; Karamouzas, Ioannis; Zordan, Victor

doi:10.1145/3359566.3360071

Cited by 8 publications

(4 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The motivation for testing spaces of different dimension ( m ) is perhaps less obvious. It has been shown repeatedly that many actions that appear complex are in fact highly coordinated (e.g., Safonova et al, 2004 ; Ranganath et al, 2019 ). It seems likely that some reduced dimensional space would be adequate for representing the entire successful action family.…”

Section: Methodsmentioning

confidence: 99%

Characterizing Continuous Manipulation Families for Dexterous Soft Robot Hands

2021

View full text Add to dashboard Cite

There has been an explosion of ideas in soft robotics over the past decade, resulting in unprecedented opportunities for end effector design. Soft robot hands offer benefits of low-cost, compliance, and customized design, with the promise of dexterity and robustness. The space of opportunities is vast and exciting. However, new tools are needed to understand the capabilities of such manipulators and to facilitate manipulation planning with soft manipulators that exhibit free-form deformations. To address this challenge, we introduce a sampling based approach to discover and model continuous families of manipulations for soft robot hands. We give an overview of the soft foam robots in production in our lab and describe novel algorithms developed to characterize manipulation families for such robots. Our approach consists of sampling a space of manipulation actions, constructing Gaussian Mixture Model representations covering successful regions, and refining the results to create continuous successful regions representing the manipulation family. The space of manipulation actions is very high dimensional; we consider models with and without dimensionality reduction and provide a rigorous approach to compare models across different dimensions by comparing coverage of an unbiased test dataset in the full dimensional parameter space. Results show that some dimensionality reduction is typically useful in populating the models, but without our technique, the amount of dimensionality reduction to use is difficult to predict ahead of time and can depend on the hand and task. The models we produce can be used to plan and carry out successful, robust manipulation actions and to compare competing robot hand designs.

show abstract

Section: Methodsmentioning

confidence: 99%

Characterizing Continuous Manipulation Families for Dexterous Soft Robot Hands

2021

View full text Add to dashboard Cite

show abstract

“…Ranganath et al. [RXKZ19] compressed the action space using principal or independent component analysis, reducing the size of the output of the DRL model. Their method is otherwise similar to [PALvdP18].…”

Section: Character Controlmentioning

confidence: 99%

“…They also sampled initial states to favour states from which the current model achieves relatively lower rewards. Ranganath et al [RXKZ19] compressed the action space using principal or independent component analysis, reducing the size of the output of the DRL model. Their method is otherwise similar to [PALvdP18].…”

Section: Other Motionsmentioning

confidence: 99%

A Survey on Deep Learning for Skeleton‐Based Human Animation

Mourot

Hoyet

Clerc³

et al. 2021

Computer Graphics Forum

View full text Add to dashboard Cite

Human character animation is often critical in entertainment content production, including video games, virtual reality or fiction films. To this end, deep neural networks drive most recent advances through deep learning (DL) and deep reinforcement learning (DRL). In this article, we propose a comprehensive survey on the state‐of‐the‐art approaches based on either DL or DRL in skeleton‐based human character animation. First, we introduce motion data representations, most common human motion datasets and how basic deep models can be enhanced to foster learning of spatial and temporal patterns in motion data. Second, we cover state‐of‐the‐art approaches divided into three large families of applications in human animation pipelines: motion synthesis, character control and motion editing. Finally, we discuss the limitations of the current state‐of‐the‐art methods based on DL and/or DRL in skeletal human character animation and possible directions of future research to alleviate current limitations and meet animators' needs.

show abstract

“…Recent DRL methods either directly imitate mocap examples [Peng et al 2018a;Won et al 2020], which makes strategy discovery hard if possible; or adopt a de novo approach with no example at all [Heess et al 2015], which often results in extremely unnatural motions for human like characters. Close in spirit to our work is [Ranganath et al 2019], where a low-dimensional PCA space learned from a single mocap trajectory is used as the action space of DeepMimic for tracking-based control. We aim to discover new strategies without tracking, and we use a large set of generic motions to deduce a task-and-strategy-independent natural pose space.…”

Section: Natural Pose Spacementioning

confidence: 99%

Discovering diverse athletic jumping strategies

et al. 2021

View full text Add to dashboard Cite

We present a framework that enables the discovery of diverse and natural-looking motion strategies for athletic skills such as the high jump. The strategies are realized as control policies for physics-based characters. Given a task objective and an initial character configuration, the combination of physics simulation and deep reinforcement learning (DRL) provides a suitable starting point for automatic control policy training. To facilitate the learning of realistic human motions, we propose a Pose Variational Autoencoder (P-VAE) to constrain the actions to a subspace of natural poses. In contrast to motion imitation methods, a rich variety of novel strategies can naturally emerge by exploring initial character states through a sample-efficient Bayesian diversity search (BDS) algorithm. A second stage of optimization that encourages novel policies can further enrich the unique strategies discovered. Our method allows for the discovery of diverse and novel strategies for athletic jumping motions such as high jumps and obstacle jumps with no motion examples and less reward engineering than prior work.

show abstract

Low Dimensional Motor Skill Learning Using Coactivation

Cited by 8 publications

References 44 publications

Characterizing Continuous Manipulation Families for Dexterous Soft Robot Hands

Characterizing Continuous Manipulation Families for Dexterous Soft Robot Hands

A Survey on Deep Learning for Skeleton‐Based Human Animation

Discovering diverse athletic jumping strategies

Contact Info

Product

Resources

About